New Morningstar PS-30M gen3 controller

mike_smike_s Posts: 9Registered Users ✭✭
Update: much of the grief described below was due to a bad firmware file which Morningstar posted on their website. They apparently posted a very old v1 firmware, marked as v5. After getting and loading a correct version of the firmware, the controller itself is behaving much better. The documentation and MSView issues described still apply, though.

(yea, I know this may sound like an unreasonable customer rant to some, but I assure you that when I began reporting these issues to Morningstar 6 weeks ago, I did so with the intention of helping them get a newly released product right. I thought they'd be very interested in bugs in a new product, before they started affecting customers. They didn't seem to care much. Everyone makes mistakes, it's how you handle them that proves your mettle. Morningstar has failed.)

Based on their reputation, I recently bought one of the new Morningstar PS-30M gen3 controllers. The new ones have programmable settings, a graphical LCD display, communications capability (Modbus, Morningstar MSView), are compact, and have a clean look. Sounds great, right? I thought so too, so got one for my RV.

I have to say, it's not nearly ready for prime time. It has significant issues, which Morningstar is aware of, and seems in no hurry to correct.

If you turn the high voltage disconnect (HVD) function off, it will act like it's charging, but isn't. It fails to do the most basic function of a charge controller. (I tried turning it off, because even if set .2V higher than the equalize voltage, it can trigger and unexpectedly turn off the load)

The charging settings on the controller don't have the range to properly support the recommended values for Trojan T-105 batteries, one of the most common deep cycle ones available.

Their MSView software doesn't support it correctly - you can't even download the controller's config, then upload it without an error. MSView hasn't been updated since January.

The documentation is at best inconsistent, and at worst simply wrong. They call the same setting by multiple names - a setting may be called one thing on the controller, a different thing in the documentation, and by two (!) other names in MSView. Good luck sorting out what's what.

For example, for the most basic setting of a charge controller, their Modbus doc calls it "Regulation voltage." The user manual calls it "Absorption stage (voltage)." MSView calls it _both_, in different places. It gets worse with the more complex settings - "Reference charge limit", "Maximum regulation limit", and "Max battery V" are used for the same thing in different places. Beyond just the settings, the functions are also confused, in some places they refer to "solar disconnect," in others "battery disconnect" for the same thing. Someone at MS must understand how these things work, but it's not the people who write the documentation. There's no consistency at all. 

Yesterday, they posted some new firmware, which they said was "v5", and which is supposed to fix at least one of the above issues, plus several others not mentioned. When I went to install it using their MSLoad software, it reported that it was "PWM Firmware update v00.0beta, 16.Feb.2016" Uh, what? I reported that to them, and they said they screwed up the build (and the QA, obviously), and later posted a version which did say it was V5. Great.

I installed the new "v5" firmware (MSLoad reported success), after which the controller reported it was running "FW:v1", and MSView said it was running "Firmware Version v01.04.01". Uh, what???

To top it off, the new firmware broke things which worked before, only making it worse - can't even upload a config created in MSView now, it stops about 1/3 of the way in with an error (worked fine before the new firmware, except for the bug mentioned above). I can only hope that it can be programmed with new firmware if and when MS gets their act together. And, the controller now has an alarm which can't be cleared. An obvious conclusion is that their developers are incompetent and working with "spaghetti code," where fixing one thing breaks another.

I don't think it's unreasonable to expect it to work correctly, after it's been in development for over a year. This should be a "set it and forget it" product. A $55 made in China Grape controller from the local Home Depot would work better. I expected more from Morningstar, based on both reputation and price.

Arrrgh. 


Comments

  • EstragonEstragon Posts: 897Registered Users ✭✭✭✭
    Is the high voltage disconnect a battery voltage, or a pv voltage? Pv voltage would likely be higher than .2 higher than Veq so a disconnect wouldn't be surprising. My guess is the HVD is there to protect the controller in case Voc gets too high with cold panels early in the morning, or maybe if used as a load controller.

    The different charge limit settings may be to limit temp compensation ranges. Firmware v1 seems like a truncation of v01.xx

    I don't work for Morningstar, but I do have some sympathy for them. Documentation issues aren't acceptable, but they happen. Engineers like to make products that work. Doc writers like to keep it simple. Sometimes stuff gets lost in translation (sometimes literally).

    There may well be issues with the product, but it sounds to me like you're now finding reasons to think every problem is a problem with the controller, not something that could be a misunderstanding of the docs. Using a pc to set custom settings can be useful, but dip switch defaults are often just fine and custom settings may hurt more than help without a clear understanding of what they do. Maybe the best thing to do would be to return the controller and get a "set it and forget it" controller?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • mike_smike_s Posts: 9Registered Users ✭✭
    edited May 19 #3
    Well, the user manual calls it "Battery / Load High Voltage disconnect (HVD)". In MSView, once you can get into their help system (you can't get to it by clicking the help button in the screen where you set these values, and they don't have any section specific to this controller), you find "High Voltage Disconnect - The battery voltage threshold at which the solar array is disconnected and charging is halted. If the battery voltage increases beyond this value, charging is stopped until battery voltage falls to a safe level." Their Modbus doc says "Disconnect the loads if the battery voltage rises too high." The user manual shows voltage levels for:

    HVD - load (@ 25ºC) 15.3V
    HVD - solar (@ 25ºC) 15.2V

    So, Morningstar is saying it's a battery, or solar, or load disconnect, depending on what you read. And they're saying it's based on battery, solar, or load voltage, depending on where you read. So they've covered all the bases. Just another example of the quality of their documentation.

    You tell me the answer to your question. My best guess is that it's based on the battery voltage, and experience shows that it disconnects the load, not the array.

    But that makes little sense - the controller is powered by the battery, so that's always connected. It can switch the solar or the load, since it must do PWM by switching the solar, and also has the ability to switch the load since it can support lighting control. The solar voltage (PWM averaged) is basically the same as the battery voltage when connected, so if that gets too high, it should just disconnect the array, not the load. If the solar voltage goes above the max while the array is disconnected, that's an implementation problem (Voc too high), there's little that can be done, and it makes no sense to disconnect the load from the battery.

    No, I don't need your condescending attitude or a dumbed down controller which won't properly charge a battery (Morningstar's DIP switch absorption setting tops out at 14.7V for FLA, short of what Trojan, or Interstate, or US Battery, or Rolls recommends). The MS controller would be great, if it did what they claim it does. But, Morningstar is just a f'd up mess.

    Oh, and "v1 seems like a truncation of v01.xx" is obvious, and doesn't explain why that's shown when firmware v5 is programmed. 5 != 1.
  • mcgivormcgivor Posts: 784Registered Users ✭✭✭✭
    edited May 19 #4
    You're dead on about the documentation, had to read everything, more than once, the Schneider manual was much simpler and straight forward, the ability to set advanced custom settings with the LCD display without having to use software was an added bonus Personally I like the Morningstar products but am impressed by the Schneider even more so. Sorry to here about your issues but your  perseverance will eventually prevail, good luck, wish I could offer a solution.
    System 1, 6×250W Schutten poly 2S/3P 24V nominal, Schneider 60-150 MPPT, 260 Ah FLA battery, Cotek 1000W 230V  PSW inverter. 
    System 2, 4× 315W Axitec poly parallel 24V nominal,  Morningstar TS 60 PWM,  440 Ah FLA battery, Armon 2000W 230V  PSW inverter.
    System 3 4× 100W Schutten mono parallel 12 nominal, Morningstar TS 60 PWM 200Ah FLA battery, Samlex 200W 120V PSW inverter.
  • mike_smike_s Posts: 9Registered Users ✭✭
    edited May 19 #5
    The thing is, the Morningstar hardware seems great - much better looking, a more compact form factor (smallest Schneider/Xantrex is 8x5x2.5", this thing is about 6x4x2), and has a very useful graphical display and on-board control/menu system. Maybe your're talking about a fancier one, my need is pretty basic for a small RV, don't need MPPT, so I looked at the C35, where you set voltages with pots. A quick look at the Schneider shows it does charging or load control, not both at once. The MS does both, and will show and track both solar and load currents, and even keep (and graph) daily records for stuff like load Ah and charge Ah.

    It has great promise, just that the current software and documentation are so bad. And Morningstar doesn't seem to care much - it took them over 6 weeks just to fix an issue where it wouldn't charge at all, while acting like it was (perhaps - their release notes don't say it's fixed, and I haven't tested it because I have a work-around configured).
  • mike_smike_s Posts: 9Registered Users ✭✭
    Not sure how, but when trying to add some info to an earlier post, the whole thing got deleted. In response to the question about how the HVD works:

    Well, the user manual calls it "Battery / Load High Voltage disconnect (HVD)". In MSView, once you can get into their help system (you can't get to it by clicking the help button in the screen where you set these values, and they don't have any section specific to this controller), you find "High Voltage Disconnect - The battery voltage threshold at which the solar array is disconnected and charging is halted. If the battery voltage increases beyond this value, charging is stopped until battery voltage falls to a safe level." Their Modbus doc says "Disconnect the loads if the battery voltage rises too high." The user manual shows voltage levels for:

    HVD - load (@ 25ºC) 15.3V
    HVD - solar (@ 25ºC) 15.2V

    So, Morningstar is saying it's a battery, or solar, or load disconnect, depending on what you read. And they're saying it's based on battery, solar, or load voltage, depending on where you read. So they've covered all the bases. Just another example of the quality of their documentation. emoticon

    My best guess is that it's based on the battery voltage, and experience shows that it disconnects the load, not the array. 

    But that makes little sense - the controller is powered by the battery, so that has to always be connected. It can switch the solar or the load, since it must do PWM by switching the solar, and also has the ability to switch the load since it can support lighting control. The solar voltage (PWM averaged) is basically the same as the battery voltage when connected, so if that gets too high, it should just disconnect the array, not the load. If the solar voltage goes above the max while the array is disconnected, that's an implementation problem (Voc too high), there's little that can be done, and it makes no sense to disconnect the load from the battery. 
  • mcgivormcgivor Posts: 784Registered Users ✭✭✭✭
    You are correct, larger units, even the MS TS60,   can't do both load control and  charge control similtaniously.
    System 1, 6×250W Schutten poly 2S/3P 24V nominal, Schneider 60-150 MPPT, 260 Ah FLA battery, Cotek 1000W 230V  PSW inverter. 
    System 2, 4× 315W Axitec poly parallel 24V nominal,  Morningstar TS 60 PWM,  440 Ah FLA battery, Armon 2000W 230V  PSW inverter.
    System 3 4× 100W Schutten mono parallel 12 nominal, Morningstar TS 60 PWM 200Ah FLA battery, Samlex 200W 120V PSW inverter.
  • EstragonEstragon Posts: 897Registered Users ✭✭✭✭
    My post was not intended to be condesending. In my experience, people sometimes get frustrated enough with a product that working with it further is counterproductive. I've reached that point myself. Your original post had that tone (eg "Morningstar lacks adult supervision").

    Looking through the manual, I agree that some things are unclear or don't make sense. The HVD is one. The manual lists a dip switch setting for L16 with Vabs (15.4v) and Veq values higher than the default HVD. HVD is apparently customizable, but it doesn't make sense to me that there should be conflicting default settings.

    Does your unit not have the L16 setting? Are there limits to the custom values that can be set for batteries needing higher voltage charging?

    Although disconnecting loads with a HVD seems odd, as solar disconnects at a lower voltage by default, there could be other charge sources that might raise voltage to load damaging levels. Are there limits to custom HVD settings?

    For the firmware version, it sounds like the update may not have worked. When rebooting, could the battery have been disconnected/reconnected while there was voltage on solar input? V1.xx to v5 seems like it might be a big jump... maybe an interim version needs to be loaded before v5? Are there features in v5 absent in v1 that might indicate which version is running independent of the version indication?
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • mike_smike_s Posts: 9Registered Users ✭✭
    According to their MSLoad documentation, you get a green checkmark if it worked, and a red X if it didn't. It indicated that it worked.

    Since the displayed version is now different, I can only believe that to be true. Before I loaded what they called "v5", the unit showed HW 1.4 SW 4. After loading it, it shows HW 1.4, SW 1. Go figure, but it seems to me very likely that they posted an old, even buggier version, and called it v5.

    That wouldn't be much of a surprise, since the first one they posted was marked (internal to the file, and displayed by MSLoad when installing) as v00.0beta. When I noticed that, I let them know. They later replaced the file with a new one, which was internally labeled as v05, and claimed that the first one was v5, just marked wrong. The replacement "v5" is what shows as FW:v1.

    I think they're just as disorganized as their documentation.

    Regarding the settings, it's my intent to upgrade to 2x Trojan T-105s, a very popular setup for RVs. Trojan's current charging recommendation calls for 14.82 Vbulk, 13.5 Vfloat, and 16.2 Veq. None of the DIP settings are even close across the board. And the original firmware wouldn't allow you to set anything above 16 V from the keypad. MSView does, but the firmware "upgrade" broke it - it won't upload a config now.
  • mike_smike_s Posts: 9Registered Users ✭✭
    Update (also added to OP): much of the grief described was due to a bad firmware file which Morningstar posted on their website. They apparently posted a very old v1 firmware, marked as v5. They themselves has issues with it when they loaded it on their own units. After getting and loading a correct version of the firmware, the controller itself is behaving much better. The documentation and MSView issues described still apply, though.
  • EstragonEstragon Posts: 897Registered Users ✭✭✭✭
    Setting flooded 6 appears to have Vabs at 14.7 and float at 13.5. To me, that's a close enough match to Trojan spec try it for a while and adjust if needed. The voltage may be a bit low vs spec, but the time, at 3hrs is fairly long IMHO.

    Veq at 15.4v might be a bit low, but eq is more of an art than a science anyway. I wouldn't set Veq (@25°C) higher than 16v unless I was literally going to sit with the batteries during the eq. I'd rather set them to a lower voltage for a longer time and check periodically than risk overheating or loss of electrolyte.

    I don't have any other ideas about the firmware.
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • davemacdavemac Posts: 39Registered Users ✭✭
    I don't have this exact Morningstar product but have had success with my settings that could be of help to you?

    I have the Morningstar TS MPPT 60 on a bank of 4 x T105 (series) for the past 12 months with max of 20% DOD.

    I opted for the factory charge controller setting - Battery Type 6 - Flooded who's specs completely match the Trojan typical charge profile for T105. After a year, specific gravity of FLA's are all up in the 1.283 range (an improvement over the past 6 months) which is ideally healthy when completely charged and rested after equalization.

    All the best with your efforts to get your system running smoothly.
Sign In or Register to comment.